View
6
Download
0
Category
Preview:
Citation preview
Working Paper 259
An Assessment of Inflation
Modelling in India
B. Karan Singh
April 2012
INDIAN COUNCIL FOR RESEARCH ON INTERNATIONAL ECONOMIC RELATIONS
Contents
Abstract .............................................................................................................................. i
1. Introduction ................................................................................................................ 1
2. Concepts and Their Measurement ........................................................................... 4
3. Measuring Inflation in India ..................................................................................... 6
4. Measuring Output Gap in India ............................................................................... 7
5. New Keynesian Phillips Curve (NKPC) and Supply Shocks ................................. 9
6. Empirical model and Results of the Phillips Curve in the case of India ............. 12
7. Evidence of Failure in Co-ordination between Fiscal and Monetary policies .... 13
8. Policy Recommendations and Conclusions ........................................................... 15
References ....................................................................................................................... 17
List of Table
Table 1. Estimated Open-EcTable 1 Estimated Open-Economy Phillips Curve for
Period 2 (Q2 2004 – Q4 2009) .......................................................................... 21
Table 2. Data Definitions and Sources ............................................................................ 22
List of Figures
Figure 1: Correlation between the Different Measures of Inflation from Q4 1996-97
to Q2 2011-12 .................................................................................................... 6
Figure 2: Inflation in CCPI (%) from Q4 1996-97 to Q2 2011-12 ................................... 7
Figure 3: Output Gap – HP Method (per cent) ................................................................. 8
Figure 4: Output Gap – Kalman Method (per cent) .......................................................... 8
Figure 5: Shape of the Phillips Curve for Demand and Supply Shocks ......................... 11
Figure 6: Relationship between Inflation and the Output Gap during Period 2 (Q1
2004-05 to Q2 2011-12) .................................................................................. 13
Figure 7: Real Interest Rates – CCPI .............................................................................. 14
Figure 8: Change in Fiscal Deficits as a Percentage of GDP .......................................... 15
List of Appendix
Appendix 1 ....................................................................................................................... 21
i
Abstract
This study analyses India’s inflation using the Phillips curve theory. To estimate an
open-economy Phillips curve, we need three variables: (1) inflation (2) the output gap
and (3) the real effective exchange rate. In India, the incorrect measurement of
variables causes much difficulty in estimating the Phillips curve. The study by Singh,
B.K. et al (2011) found that the Phillips curve existed, after addressing the issues
related to measuring the variables. They suggested that the composite consumer price
index (CCPI) was the best measure of inflation, and should be used to construct the
real effective exchange rate in India. In measuring the output gap, the paper found that
the Kalman filter estimates of the output gap capture all the dynamics of the Indian
economy. This study constructs the variables, using a method similar to that followed
by Singh, B.K. et al (2011). During the period 2008 to 2011, a combination of adverse
supply shocks and overheating of economy resulted in an inflationary situation.
____________________
JEL Classification: C32; E31; E50
Keywords: Kalman Filter, Output Gap, Inflation, Phillips Curve
Authors’ Email Address: karan@icrier.res.in
__________________
Disclaimer:
Opinions and recommendations in the paper are exclusively of the author(s) and not
of any other individual or institution including ICRIER.
1
An Assessment of Inflation Modelling in India*
B. Karan Singh
1. Introduction**
One of the major pressures on India’s macroeconomic climate following the global
financial crisis is an uncomfortably high level of inflation. Between the period 2008-09
and 2011-12, inflation in India can be attributed to both microeconomic and
macroeconomic factors. The microeconomic factors are primarily supply side
problems, whereas the macroeconomic factors relate to the effectiveness of demand
management through fiscal and monetary policies. Supply side problems refer to
adverse supply shocks or cost-push inflation. One of the main sources of the adverse
supply shock was the failure of the kharif season1 production in 2008-09 and 2009-10.
The other source was the high volatility in the prices of selected commodities in the
global market, mostly fuel and basic metals.
Demand-side management had been a challenge during this period, not only because of
the adverse spill over effect of global economic uncertainties on the Indian economy
but also because of the political cycle (the general elections were held in 2009) that led
to unsustainable fiscal expansion. During the financial years 2008-09 and 2011-12,
both the fiscal and monetary stances were loose and there was little or no indication
that these would be tightened.
In 2008-09, the need to stimulate the economy following the global financial crisis
dictated the stance on fiscal and monetary policy. Although there was co-ordination of
monetary and fiscal policies during this phase, the co-ordination weakened when the
government sought to exit the fiscal stimulus. After 2008-09, fiscal consolidation was
* I am thankful to Dr. Parthasarathi Shome for his continuous support and guidance in writing this paper.
I am grateful to Dr. Shankar Acharya and Dr. Deena Khatkhate for their review and useful comments.
A previous version of the paper was presented at the Konrad-Adenauer-Stiftung (KAS) 3rd quarterly
seminar series on macro financial sector, 2011, in which Chairperson Dr. Surjit Bhalla, and Mr. N.K.
Singh, Dr. Sanjaya Panth, Dr. Pooja Sharma and other participants gave valuable comments for which
I am thankful. I further extend my gratitude to my colleagues and friends, Dr. Nisha Taneja, Dr. Saon
Ray, Dr. Vijay Kumar Varadi, Dr. Ranjan Kumar Das, Dr. Jyotirmoy Bhattacharya, Dr. Shalini Bhatia,
Dr. T.A. Ananth, Ms. Asha Mukherjee, Mr. Swaraj Mehta, Ms. Tara Nair, Mr. A. Kanakaraj and Ms.
T.O. Sridevi for their help and support at various stages of the preparation of this paper.
** This study is analyses recent inflation in India. It is an extension of my previous paper co-authored
with A. Kanakaraj & T.O. Sridevi, which was published in the Journal of Asian Economics. The
extension of the paper lies in the analysis of the interaction between supply shocks and demand
management policies.
1 Kharif season production refers to the months between October and December. The importance of this
season is that during this season, the country harvests a maximum proportion of its food grains.
2
achieved at the state and central government levels in 2010-11 but the tightening of
monetary policy was more gradual. Policy rates were raised several times in tiny doses
that kept the real interest rate down; the interest rate was negative2 in 2010-11. In 2011-
12, though the real interest rate finally reached a positive level in a few months, the
slippage in fiscal consolidation proves that the fiscal-monetary policy co-ordination
was a failure.
What should be the monetary policy action when an economy is facing supply-side
shocks?
The study by Rasche and Tatom (1981), which is based on international evidence,
argues that though supply shocks are temporary in nature, monetary tightening is
required. Blanchard and Gali (2007) found that improvements in monetary policy
reduced the impact of the 2000s oil shock on the US economy. However, many
monetary economists believe that for monetary policy considerations, core inflation3
that excludes supply shock components should be used rather than headline inflation. It
is believed that the removal of volatile and transient components (i.e. exclusion of
supply shock components) from headline inflation can predict the monetary
phenomenon of inflation. But Walsh (2011) found that in developing economies,
shocks to food inflation transmitted to non-food inflation. He, therefore, argues that the
use of core inflation as a measure of inflation is subject to severe bias. Blanchard,
Ariccia and Mauro (2010) contend that inflation expectation could be anchored well if
central banks target headline inflation rather than core inflation.
Studies by Rashes and Tatum (1981), Blanchard and Gail (2007) and Walsh (2011)
found that if an economy is characterised by poor macroeconomic demand
management, the effects of transient supply shocks would continue for longer than the
time it takes for the shock to wear out, and create considerable inflationary pressures.
India’s recent inflationary situation suggests that the supply shock to inflation might
have been poorly managed. Between Q1 2006-07 and Q2 2011-12, India’s Consumer
Price Index (CPI) based inflation was above 6 per cent. During this period, India also
experienced adverse supply shocks to inflation. Reasons for the supply shocks, and
what the government should have done in these circumstances, was widely discussed.
Kaushik (2011a) argues that food inflation has been managed poorly through
misguided government intervention in the food grains market. Kaushik (2011b) argues
that focusing on removing supply side bottlenecks and adopting policies that would
improve productivity would dampen inflation.
Apart from studies on supply side issues, differing views have also been expressed on
the role of macroeconomic demand management to achieve a stable, low level of
2 A negative real interest rate could attract investors to invest in commodities rather than in productive
investments. As a result, it tends to fuel inflation further.
3 Core inflation is a measure of price rise in the case of non-fuel and food commodities.
3
inflation. The report by Rajan (2009) on financial sector reforms recommended that the
RBI should focus on the single objective of inflation control, i.e., it should try to
maintain inflation close to a targeted low rate or to stay within a given range. The
report concludes that by doing so, the RBI can achieve stability in growth as well as in
inflation. Acharya (2009), in a counterargument, contends that given the frequent
occurrence of supply shocks in the Indian economy, the single objective of inflation
control is not the right choice for the RBI at this stage of development. Inflation due to
supply side shocks proves a monetary response to be ineffective, and in such cases, the
RBI would have to opt for a trade-off between inflation and growth.
There has been no empirical evaluation based on theoretical foundations of the recent
inflation situation in India. RBI policy papers do not provide enough of a theoretical
explanation for its recent policy actions. The possible reasons for the seeming inability
of the RBI to contain the inflationary pressure are 1) divergence in inflation trends
reflected by the WPI and CPI, which leads to misreading of the seriousness of the
situation 2) the belief that monetary policy is not effective in controlling food inflation
3) global economic uncertainties and 4) the relatively lower GDP growth in India as
compared to the pre-crisis level.
One of the most widely used theoretical models in monetary policymaking is the theory
of the Phillips curve. The original theory of the Phillips curve published in 1958 by
A.W. Phillip has been enriched over the years. The empirical relationship between
inflation and the output gap – that is the Phillips curve – could guide monetary policy to
achieve low and stable levels of inflation. Older Keynesians define the output gap as
arising primarily due to inflationary pressures, though the New Keynesian Dynamic
Stochastic General Equilibrium theory posits that the output gap arises primarily due to
nominal rigidities. The classical model4 by Lucas (1973) concludes that the output gap-
inflation association is entirely contemporaneous. He also points to the existence of
lagged effects, which are difficult to explore in a short-term time series. The paper
concludes that the simple structure of the relationship between inflation and the output
gap captures the main phenomenon predicted by the natural rate theory. To take
globalisation into account, Ball (1998) added the real effective exchange rate in the
Phillips curve equation. Gordon (2011) says that the contribution of post-Keynesian
theory in the Phillips curve model is that inflation and output gap is no longer
positively correlated. It could have any correlation depending on the relative
importance of aggregate supply and demand shock.
4 The model posits that the quantity supplied in each market will be determined by a normal component
and a cyclical component, which vary from market to market. The model defines the output index in
each market as z, and uses ynt and yct to denote the logs of the two components. In the model, the
supply in market z is given as yt(z) = ynt + yct(z), where ynt is the normal component, which reflects
capital accumulation and population change, and yct is the cyclical component, which varies with
perceived relative prices and also with its own lagged value.
4
How well the theory of Phillips curve has been used in monetary policymaking is
analysed by Meade and Thornton (2010). They analysed the Federal Open Market
Committee transcript to see how intensively the Phillips curve framework had been
used in the policymaking process. They counted the keywords of the Phillips curve
theory in the official transcript. The keywords were potential output, output gap and
Phillips curve.
The RBI has also published papers on estimating the potential output (Bordoloi, Das &
Jangili, 2009) and estimating the threshold inflation using the backward-looking
Phillips curve framework (Mohanty, Chakraborty, Das & John, 2011). But the RBI did
not use these keywords in its policy statement.
The purpose of this paper is to discuss the recent developments in estimating the
Phillips curve in the Indian context. The remaining portion of this paper is structured as
follows. Section 2 reviews the estimation of inflation and the output gap in the Indian
context. Sections 3 and 4 discuss the best methods to measure inflation and estimate the
output gap, respectively. Section 5 discusses the theory of the Phillips curve with
reference to supply shocks. Section 6 discusses the existence and shape of the Phillips
curve in India. Section 7 discusses the role of monetary and fiscal policies in containing
inflation in recent times. Besides summarising the main arguments, the last section also
outlines some policy recommendations.
2. Concepts and Their Measurement
The Phillips curve predicates the relationship between inflation and the output gap.
Thus, it is critical to define these concepts and examine how they are measured.
Output Gap: The output gap is a measure of the gap between the actual output of an
economy and the potential output it could achieve when it is operating at full capacity.
In demand-side driven economies, a positive value for the output gap indicates
inflationary pressure in the economy, i.e. it is operating at a higher marginal cost. In
supply-side driven economies, a positive value for the output gap indicates a
deflationary situation in the economy, i.e. it is operating at a lower marginal cost. In
general, the output gap is measured on two scales – the deviation of the actual level of
output from the potential output level, and the growth of output relative to potential
growth (the latter is called ‘speed limit policy’). In India, the RBI announces forecasts
of GDP growth while designing the monetary policy. Since the RBI does not announce
its forecast of the GDP level, it is possible to follow the speed limit policy.
This is not the first attempt to measure the output gap for the Indian economy. There
have been several studies in this regard - Callen and Chang (1999), Ray and Chatterjee
(2001), Kundan (2009) and Paul (2009), Srinivasan (2009), Dua and Gaur (2009).
These studies suffer from drawbacks. One is that they use the index of industrial
production to estimate the output-gap. However, since the contribution of IIP to the
5
total GDP is only 16 per cent (Central Statistical Organisation, India – National
Accounts Statistics, 2012), using IIP as a proxy for growth allows at best a partial
analysis. To overcome this, economists should use quarterly GDP data.
The second problem relates to the use of the HP model to estimate the output.gap.
Ozbekand Ozlale (2005) points out that the HP filter cannot capture the excessive
boom-and-bust cycle, which is a well-known characteristic of many emerging markets.
Another serious problem with the HP method is that it defines the estimated smoothed
series of GDP as potential output. Basu and Fernald (2009) say that so far, no macro
economic theory has proven that potential GDP is a smoothed series. The present paper
also contends that the HP method is inappropriate in the context of the Indian economy.
Virmani (2004) estimated the output gap using an unobserved components model for
the Indian economy. This model is an extended version of the Kalman model. Although
in the original estimation of the model, Kuttner (1994) used the consumer price index
(CPI) as a measure of inflation to estimate the forward-looking Phillips curve, Virmani
chose the wholesale price index (WPI).
Inflation: Inflation is defined as the rate of increase in the index of general prices,
which is the other variable in the Phillips curve. Both the wholesale and consumer price
index have been used as measures of inflation in economic literature. It is important to
take into account the limitations of both the CPI and the WPI as measures of inflation
while estimating the Phillips curve. Srinivasan T.N. (2008) describes the shortcomings
of using the WPI as a measure of general inflation in India. The WPI is constructed for
a given basket of goods in the economy. Data on prices comes from various sources –
the farm gate, the factory gate, primary markets, secondary markets, wholesale markets
and retail markets. The WPI, therefore, captures neither the supply side (producer
price) nor the demand side (market price). On the other hand, Shapiro and Wilcox
(1996) found that the CPI tends to overestimate inflation by 0.6 to 1.5 percentage points
per year. This raises serious doubts about the appropriateness of using the CPI as a
measure of inflation. The Boskin Commission Report (1996) also found a similar
upward bias in CPI estimation, ranging between 0.8 to 1.6 percentage points per year.
However, Shapiro and Wilcox (1996) conclude that the inaccurate measurement of the
CPI has no implications for the output gap or natural employment. Technically, the CPI
captures both demand side factors as well as agents’ expectations.
In the Indian context, the reliability of the CPI as a measure of inflation can be
questioned. For instance, the base year for calculating the CPI for agricultural and rural
labourers is still 1986-87. Therefore, it is unlikely to capture the price rise in items that
have entered the consumption basket subsequently for the two income groups. Given
that the structure of the Indian economy has been changing rapidly, the assumption that
the consumption basket has remained unchanged over the last two decades raises
serious questions about the reliability of the statistics. In this paper, it is assumed that
continuing to use the old basket of products does not lead to time-varying biases. That
6
is, it is assumed the bias because of over or under estimation of inflation, because of the
assumption that there is no change in the consumption basket, is assumed to be constant
over the years. Further, since there is no better alternative to measure inflation, this
paper uses the CPI as a measure of overall prices in the economy.
3. Measuring Inflation in India
For an economically diverse country like India, a single measure of inflation does not
suffice. At present, India has three consumer price indices that account for three diverse
groups, viz., agricultural labourers, rural labourers and industrial workers. Besides
these, the wholesale price index is also used to measure inflation. The issue that arises
is which of these four indices should be used by the RBI to formulate monetary policy.
In the beginning of the nineties, central banks used the WPI as a measure of inflation.
In recent years, however, many central banks have given up using the WPI, as it is not
deemed an accurate measure of general inflation levels. Khatkhate (2006) presents a
case for an aggregate measure based on the CPI for the country as a whole, which could
then be used to design monetary policy. Although the government has come up with a
CPI for the country as a whole, the data is available only from 2010. Hence, in this
study, as an alternative, a Composite CPI (CCPI) has been constructed using CPI for
agricultural labourers, rural labourers and industrial workers. Each income group
carries equal weight in the CCPI. To check whether the constructed CCPI captures
overall inflation the four measures of inflation - CCPI, WPI, the GDP deflator, and
private final consumption (PFC) deflator - have been correlated. Figure 1 presents the
correlation graph of the different measures of inflation.
Figure 1: Correlation between the Different Measures of Inflation from
Q4 1996-97 to Q2 2011-12
Source: Author’s own calculations using the Online Database on the Indian Economy, Reserve
Bank of India
7
As expected, the correlation between WPI and CCPI is nowhere near perfect. But the
constructed CCPI and PFC deflator have correlated well. Since the PFC deflator is
released with huge lags as compared to the CCPI, throughout this paper, the CCPI has
been used as a measure of inflation.
Figure 2 shows that inflation based on the CCPI is above 6 per cent since Q1 2006-07,
the period after the Indian economy saw significant adverse supply shocks. The shock
pertains to food and fuel.
Figure 2: Inflation in CCPI (%) from Q4 1996-97 to Q2 2011-12
Source: Author’s own calculations
4. Measuring Output Gap in India
The output gap is an unknown variable, which helps policymakers decide whether the
economy is operating below its potential output or above it. There are a number of
methods to estimate the output gap, although there is no theoretical consensus on how
the gap should be estimated. In empirical macro modelling, the output gap is a crucial
and basic variable. While there is considerable research on estimating the output gap, in
practice economists use various methods of estimation. Two criteria have been used in
this paper to evaluate the output gap: 1) how well it captures major economic events
and 2) how well it explains inflation dynamics. As stated earlier, a number of studies
have estimated the output gap for India by using the HP method, which does not yield
reliable output gap estimates. Figure 3 presents evidence for the proposition that the HP
method has severe limitations due to its smoothing assumptions. As a result, between
Q3 2003-04 and Q3 2005-06, the HP method overestimates the potential GDP growth,
while it underestimates the potential growth for the period between Q1 2006-07 and Q4
2007-08.
-2
0
2
4
6
8
10
12
14
16
18
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12
Infl
ati
on
(%
)
Quarter
8
Figure 3: Output Gap – HP Method (per cent)
Source: Author’s own calculations
These results are inconsistent with India’s actual growth trajectory because the Indian
economy is in transition and its growth is not a smooth series like that of a developed
country. Hence, the application of the HP method in the context of developing
countries like India does not reveal the changes in the economy. Figure 4 shows that the
Kalman filter estimates that accurately capture important economic events in the Indian
economy.
Figure 4: Output Gap – Kalman Method (per cent)
Source: Author’s own calculations
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11
Ou
tpu
t G
ap
(%
)
Quarter
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12
Ou
tpet
Ga
p (
%)
Quarter
9
As Figure 4 shows, the trend in the output gap during Q4 1996-97 and Q2 2003-04 was
dynamic, and recorded very high levels of deviation. The prime factors behind this
instability were the Asian financial crisis, the dotcom bubble burst, monsoon failures, a
structural break as a result of the investment boom and a shock in the services sector.
From Q4 2002-03 to Q2 2003-04, the output gap was recorded at an average of above
1.7 per cent because investment rates rose by 4 per cent of the GDP, which was a
significant structural break in India’s growth history. This has been clearly captured by
the Kalman method. The HP method, however, fails to capture this, because of which
the estimates for the following periods are inconsistent. After the structural break, the
Indian economy entered a period in which the country achieved its highest consecutive
five-year average growth since independence, i.e. 8.9 per cent between 2003-04 and
2007-08. Surprisingly, the output gap for the period following the structural break and
until the global crisis impacted the Indian economy fell substantially. From Q1 2004-05
to Q3 2007-08, the output gap fell in the range of –1 and +1 per cent. The global
financial crisis footprint on the Indian economy is captured in the revised version of the
quarterly data based on 2004-05 as the base year, whereas in Q4 2007-08, the output
gap was recorded at -1 per cent. In Q2 2009-10, the output gap reached 1.7 per cent due
to the extended second round of fiscal stimulus, one of the main components of which
was the Sixth Pay Commission, which raised the salaries of the central government
employees substantially. In the following quarter, Q3 2009-10, the Indian economy
witnessed a monsoon failure, and the output gap reached a level of -1.4 per cent.
During Q3 2010-11, the output gap fell to -1.1 per cent due a shock in the non-
agricultural sector. This could have been because of rising input costs, and a spill over
of the slowdown in manufacturing to the services sector through the channel of inter-
industry linkage. In the first half of 2011-12, the output gap values were negative.
5. New Keynesian Phillips Curve (NKPC) and Supply Shocks
Equations (1), (2) and (3) have been taken from Woodford (2003), which states the
theoretical relationship between inflation and the output gap.
(1)
where
= Inflation at time t
= Actual output at time t
= Potential or natural output at time t
1)ˆˆ( tt
n
ttt EYY
t
Y tˆ
Yn
tˆ
10
1−ttE π = Expected inflation for the time period t+1 at the time t
The reduced form of the New Keynesian Phillips Curve is
(2)
where
= Output gap
Woodford (2003) argued that in special circumstances, an economy is likely to face
exogenous shocks.5 However, equation 2 does not account for this. So equation 2 is
transformed into equation 3, taking into account exogenous shocks occurring in the
economy by adding an additional variable .
(3)
where
= Exogenous shock at time t
In early literature, Poole (1970) argued that the monetary policy objective of
stabilisation should focus on both inflation and output. According to Woodford (2003),
recent literature has arrived at a consensus that the objective of central banks should be
to move away from output stabilisation and towards the output gap. From an efficiency
point of view, and in normal circumstances, the role of output stabilisation means that
the central bank intervenes only to limit the output gap. Since the output gap is a short-
term deviation from natural output, an important role of monetary policy is to focus on
the fall and rise in the temporal component of output.
On the one hand, this may lead to the conclusion that the target output gap should be
zero at every point of time. On the other hand, in theory, according to equation 2,
complete stabilisation of the output gap and future expected inflation should lead to
lower inflation. But Woodford (2003) argued that equation 2 does not hold true in all
circumstances. In special circumstances, the residual term , added in equation 2,
might become critical due to exogenous shocks faced by the economy. As a result,
equation 3 would be the appropriate model to explain the structure of the economy in
5 Exogenous shocks arise from supply channels. Fall and rise in supply are called negative supply shocks
and positive supply shocks respectively.
πβκπ 1++= tttt Ex
YYxn
tttˆˆ −=
µt
µπβκπ ttt
n
ttt EYY ++−= +1)ˆˆ(
µt
µt
11
the presence of random shocks such as cost-push inflation and a zero lower bound on
nominal interest rates. This is why, in practice, central banks have discretionary
policies to decide the trade-offs between inflation stabilisation and output stabilisation.
In the Indian context, it would be interesting to examine whether there is an empirical
relationship between inflation and the output gap, as stated by equation 2. Goyal and
Pujari (2004) have found that the Indian economy is subject to large supply shocks.
Inflation in India may be high due to negative supply shocks or because of the residual
term. These negative shocks come from two main sources – the failure of monsoon, and
high volatility in global crude oil prices.
Most macroeconomic textbooks discuss how the relation between the price level and
output is affected by demand shocks and supply shocks. Here, however, the
relationship between inflation and the output gap has been examined.
Figure 5: Shape of the Phillips Curve for Demand and Supply Shocks
Demand Shock Supply Shock
The Phillips curve is based on the assumption that the aggregate demand curve (AD) is
sloping downward and the aggregate supply curve (AS) is moving up. In the graph
above, the initial price level is P, and its corresponding output level is the potential
output. In the demand shock scenario, two positive demand shocks have been assumed,
which result in the demand curve shifting outwards to the right from AD1 to AD2 and
then to AD3. One negative shock has also been assumed, with the demand curve
Yn
ˆ
12
shifting inwards from AD1 to AD0. The positive demand shocks raise the price level
by a magnitude of ‘a’ in one case and by ‘2a’ in the other case (where a>0). The
negative demand shock results in a decline in the price level by ‘–a’. Based on the
interaction points between the aggregate demand curve and aggregate supply curve, the
equilibrium outcomes of the different situations have been mapped. Since the objective
is to derive the Phillips curve, the equilibrium price level to inflation ‘π’ has been
rearranged. Here inflation is defined as the percentage deviation of the equilibrium
price level (points P - a, P, P +a. P+ 2a) from the initial price level P. The output gap is
equal to the percentage deviation of the actual output ( ) from the potential output
(( - )/ ). In the case of a demand shock, mapping the relationship between
inflation and the output gap shows a positive relationship exists between them. In the
case of a supply shock of the same magnitude, the resulting relationship is negative. As
a result, the central banks’ job is easy whenever there is a demand shock. The central
banks can stabilise the economy and bring it to a more efficient equilibrium with low
inflation and an optimal level of output, by targeting the output gap level at zero. But
for supply shocks, monetary policies can only target non-zero output gap levels. It is
here that a trade-off between inflation and the output gap needs to be considered when
formulating monetary policies. The optimality condition in choosing between the two
goals depends upon the relative weights of the goals. In theory, the weights of the two
variables are derived from a loss function (Woodford, 2003).
6. Empirical model and Results of the Phillips Curve in the case of India
The study by Singh, B.K. et al (2011) found that correcting the measure of inflation and
the output gap by the CCPI and the Kalman filter proves the existence of the open
economy Phillips curve in India (refer Appendix A for the detailed regression results).
This study has used the key finding of their paper. The main findings are as follows
the Phillips curve exists during the period after Q1 2004-05, the Phillips curve exists
only after controlling for supply shocks and the real effective exchange rate does not
affect the CCPI inflation. Supply shock periods are those in which agricultural
production declines and the year-on-year change in fuel prices is above 15 per cent or
below 0 per cent. Figure 7 presents the linear relationship between the output gap and
inflation. As proposed by the theory, a positive relationship exists for the demand
shock, and a negative relationship occurs for the supply shock. The positive
relationship between inflation and the output gap shows that demand-driven inflation
exists in the economy. Hence, policymakers have the option of bringing down the
inflation level by sacrificing some level of output. How much of the output is to be
sacrificed depends upon the slope coefficient. Cross-country evidence indicates that the
slope co-efficient or sacrifice ratio is subject to changes (Anderson and Wascher 1999).
They found the sacrifice ratio is smaller for countries that target inflation explicitly.
After the financial crisis, 2008-09 to 2011-12, the magnitude of the output gap in India
was not at its minimum; at the same time, the inflation rate was also high. Available
Y
Y Ynˆ Y
nˆ
13
evidence supports that the government missed the macroeconomic objectives of output
gap and inflation stabilisation.
It would be worth the attempt to study what happened during this period. However,
limitations in the quarterly GDP data pose problems in estimating key parameters. The
quarterly GDP for the base year 2004-05 has been revised twice. This was because of a
revision in the wholesale price index and the IIP, for both of which the base year was
changed to 2004-05. But the government has released GDP based on the revised IIP
series only for two years (2009-10 and 2010-11).
Figure 6: Relationship between Inflation and the Output Gap during Period 2
(Q1 2004-05 to Q2 2011-12)
Demand Shocks Supply Shocks
Source: Author’s own calculations
7. Evidence of Failure in Co-ordination between Fiscal and Monetary policies
In the year 2008-09, the global financial crisis created turmoil in the Indian economy
and worsened the slowdown in the domestic economy. As a result, the government
introduced expansionary monetary and fiscal policies.
14
Figure 7: Real Interest Rates – CCPI
Note: 1. PLR – Prime Lending Rates for five major banks.
2. Deposit Rates – Rates on deposits of more than one year maturity
Source: Author’s own calculation, using the RBI online database on the Indian economy.
On the monetary side, the real prime lending rate was brought down by 300 basis points
(see figure 8), while on the fiscal side, the fiscal deficit of the central government and
state governments was raised by 3.5 per cent and 1.1 per cent of the GDP respectively
(see Figure 9). The central government’s fiscal stimulus was stronger partly because the
slowdown occurred just before general elections to Parliament. Since a monetary policy
makes its effects felt with a lag, the monetary policy of 2008-09 would have had an
impact on the growth of the following year. As a result, 2008-09 did not witness a
problem of overheating. However, in Q3 2008-09, there was an adverse supply shock
because of the failure of food grains production in the kharif season, which caused
inflationary pressure.
-12
-10
-8
-6
-4
-2
0
2
4
6
8
10
12
Rea
l In
tere
st R
ate
s (%
)
Month
PLR – CCPI Deposit Rates – CCPI
15
Figure 8: Change in Fiscal Deficits as a Percentage of GDP
Source: Macroeconomic and Monetary Developments (July , 2011), RBI
Inflationary pressures were much more visible in 2009-10 when the economy began to
overheat because of the impact of monetary policies and the extended second round of
fiscal stimulus. The output gaps for Q2 and Q4 2009-10 touched 1.7 per cent and 1.2
per cent respectively. This was worsened by an adverse supply shock arising from a
second successive year of kharif crop failure in the third quarter of the year. The crop
failure was more severe than that of 2008-09.
In 2010-11, fiscal tightening happened at both the central and the state level, but
overheating continued because of the impact of monetary policy – real prime lending
rates had been negative from June 2009 to September 2010 after which they began
moving closer to zero. In the first half of the 2011-12, real prime lending rates reached
a positive level, but the central government’s failed attempt at fiscal consolidation for a
second year has caused overheating of the economy.
The post-crisis policy environment suggests that the supply shock was handled
inappropriately. Fiscal and monetary stimuli were not designed to achieve
macroeconomic stability. While fiscal tightening was hindered by the impending
elections, there was also a delay in tightening the monetary policy.
8. Policy Recommendations and Conclusions
The post-financial crisis period saw the government under immense pressure to tackle
the inflationary situation while simultaneously facing the challenge of boosting the
economy. The government could not progress on either front. This can be attributed to
a possible misreading of the situation because of infirmities in the data used to arrive at
-2
-1
0
1
2
3
4
2008-09 2009-2010 2010-11 2011-12 RE
Central Govt.
State Govt.
16
policy decisions. There are three major areas in which data infirmities are reflected. (1)
There is a discrepancy in inflation as reflected in the WPI and CPI. The RBI should
discontinue using WPI as a measure of inflation; a better way forward would be to
replace WPI with Producers Prices Index (PPI), which captures prices of both goods
and services and is more representative for supply side management. For monetary
policymaking, which is essentially demand management, global experience suggests
that countries use mostly CPI. It captures the agents’ expectation and demand channel.
(2) The second problem is the use of IIP as a proxy to measure overall economic
activity. Even though IIP captures only 16 per cent of the contribution to total GDP, it
has been frequently used in macro modelling merely because IIP data is available on a
monthly basis. (3) Quarterly estimates of GDP, particularly in the post-crisis period,
have proved unreliable. This data has been revised a number of times in the post-crisis
period. Strengthening the statistical capacity to ensure more reliable estimates of
quarterly GDP data will help improve policy formulation.
The consensus in recent literature is that stabilisation of the output gap helps lower
inflation and hence, should be the major macroeconomic policy objective. However,
the output gap is an unknown variable. Various methods have been suggested to
estimate the output gap, of which, this paper found, the Kalman filter estimates yield an
accurate result in the Indian context. There is still inadequate research on the estimation
of the output gap in the Indian context. Further research on this will help calibrate
policymaking more accurately.
Inflation in the post-financial crisis period was created by both supply and demand
shocks. The inability to co-ordinate fiscal and monetary policies resulted in the failure
of demand management. On the fiscal side, political constraints imposed by looming
elections impeded fiscal consolidation while the tiny incremental steps towards
monetary tightening by the RBI failed to have an impact on demand.
There was a failure to tackle the supply shocks as well. Monetary policy can be used
effectively to dampen the effects of a supply side shock. Economic literature suggests
that countries learn from earlier supply side shocks. Clearly, therefore, estimating the
contribution of supply shocks to inflation will prove a useful instrument in
policymaking. However, as this paper shows, such estimates need to look at the lagged
effects of a supply shock since estimating it on the assumption of a simple
contemporaneous relationship between inflation and the output gap would result in
inaccurate estimates. Further research on the contribution of supply shocks to inflation
in the recent post-crisis period is possible once the fully revised quarterly GDP data are
released.
17
References
Andersen, P.S. and Wascher, W.L. (1999), “Sacrifice ratios and the conduct of
monetary policy in conditions of low inflation”, Working Paper No. 82, Bank
for International Settlements, Basel, November.
Ball, L. (1998), “Policy rules for open economies”, Working Paper No. 6760, National
Bureau of Economic Research, Cambridge, January.
Basu, K. (2011a), “India’s food grain policy: An economic theory perspective”,
Economic and Political Weekly, Vol. 46 No. 5, pp. 37-45.
Basu, K. (2011b), “Understanding inflation and controlling it”, Economic and Political
Weekly, Vol. 46 No. 41, pp. 50-64.
Basu. S., Fernald , J.G. (2009), “What do we know (and not know) about potential
output?” Federal Reserve Bank of St. Louis Review, Vol. 91 No. 4, pp. 187-213.
Blanchard, O.J., Ariccia, G.D. & Mauro, P. (2010), "Rethinking macroeconomic
policy," Journal of Money, Credit and Banking”, Vol. 42 No. 1, pp. 199-215
Blanchard, O.J & Gali, J. (2007), "The macroeconomic effects of oil Shocks: Why
are the 2000s so different from the 1970s?", Working Papers No.13368,
National Bureau of Economic Research, Cambridge, September.
Bordoloi, S., Das, A. & Jangili, R. (2009), “Estimation of potential output in India”
Reserve Bank of India Occasional Papers, Vol. 30 No. 2, pp 37-73
Boskin, M. J., Dulberger, E.R., Gordon, R.J., Griliches, Z. and Jorgenson, D.,
(1996). Toward a more accurate measure of the cost of living. Final Report to
the Senate Finance Committee.
Callen, T. and Chang, D. (1999). “Modelling and forecasting inflation in India”.
Working Paper No. 119. International Monetary Fund, Washington, 01
September.
Central Statistical Organisation (2010). National Accounts Statistics 2009, New
Delhi, India
Clarida, R., Gali’, J., Gertler, M. (1999). “The Science of monetary policy: A new
Keynesian perspective”, Journal of Economic Literature, Vol. 37 No 4, pp.
1661-1707.
18
Dua, P. and Gaur, U. (2009). “Determination of inflation in an open economy
Phillips curve framework”, Working Paper No. 178, Centre for Development
Economics, New Delhi,
Gordon, R.J. (2011), "The history of the Phillips curve: Consensus and bifurcation,"
Economica, , Vol. 78 No. 309, pp. 10-50.
Goyal, A. and Pujari, A.K. 2004. “Identifying long run supply curve in India”.
Available at SSRN: http://ssrn.com/abstract=609323 (accessed 9 April 2012).
Ahluwalia, I. (1979), “An analysis of price and output behavior in the Indian economy:
1951–1973”, Journal of Development Economics, Vol. 6 No.3, pp. 353-390.
Kalman, R.E. (1960), “A new approach to linear filter and prediction problems”,
Journal of Economic Engineering, Vol. 82 No.4, pp. 34-45.
Khatkhate, D., (2006), “Inflation targeting: Much ado about something”. Economic
and Political Weekly, Vol. 41 No. 49, pp. 5031-5033.
Kundan, K.N.,(2009). Modelling inflation in India: The role of money. Munich
Personal RePEc Archive. Paper No. 16098. http://mpra.ub.uni-
muenchen.de/16098/ (accessed 9 April 2012).
Kuttner, K.N., (1994), Estimating potential output as a latent variable. Journal of
Business & Economic Statistics, Vol. 12 No. 4, pp. 326-334.
Lucas, Jr. R.E.,(1973), “Some international evidence on output-inflation tradeoffs”,
The American Economic Review, Vol. 63 No. 3, pp. 361-368.
NSSO (2004-05), Employment-Unemployment NSS 61th Round, Report no. 515 Part
I, National Sample Survey Organisation, India.
Meade, E.E. and Thornton, D.L, (2010). "The Phillips curve and US monetary
policy: what the FOMC transcripts tell us," Working Paper No. 2010-017,
Federal Reserve Bank of St. Louis, June.
Mohanty, D., Chakraborty, A.B., Das, A. and John, J. (2011), Inflation threshold in
India: An empirical investigation” Working Paper No. 18/2011, Reserve Bank
of India, New Delhi, September.
Mishkin, F.S. (2008), “Does Stabilizing Inflation Contribute To Stabilizing Economic
Activity?”, Working Paper No. 13970. National Bureau of Economic Research,
Cambridge, May.
19
Ozbek, L., Ozlale, U., (2005), “Employing the extended Kalman filter in measuring
the output gap”, Journal of Economic Dynamics & Control, Vol. 2 No. 9, pp.
1611-1622.
Paul, B.P.,(2009), “In search of the Phillips curve for India” Journal of Asian
Economics, Vol. 20 No. 4, pp. 479-488.
Poole, W., (1970), “Optimal choice of monetary policy instrument in a simple
stochastic macro model”, Quarterly Journal of Economics, Vol. 84 No. 2, pp.
197 - 216.
Rajan, R., (2009), Report on financial sector reforms. Planning Commission, New
Delhi, India.
Rasche, Robert H. & Tatom, John A., (1981), "Energy price shocks, aggregate
supply and monetary policy: The theory and the international evidence,"
Carnegie-Rochester Conference Series on Public Policy, Vol. 14 No. 1, pp. 9-
93,
Ray, P., Chatterjee, S. (2001). “The role of asset prices in India: Inflation in
recent years: Some conjectures” BIS Paper No. 8, Bank for International
Settlements. Basel.
Shapiro, M.D. and Wilcox, D.W. (1997). “Mismeasurement in the consumer price
Index: An evaluation”, Working paper No. 5590. National Bureau of Economic
Research, Cambridge, May.
Singh, B.K., Kanakraj, A. and Sridevi, T.O., (2011), “Revisiting the Empirical
Existence of the Phillips Curve for India”, Journal of Asian Economics, Vol. 22
No. 4, pp. 247-258.
Srinivasan, N., Jain, S., and Ramachandran, M., (2009). “Monetary policy and the
behavior of inflation in India: Is there a need for institutional reform”, Journal
of Asian Economics, Vol. 20 No. 1, pp. 13 - 24.
Srinivasan, T.N., (2008), “Price indices and inflation rates”, Economic & Political
Weekly, Vol. 43 Vol. 26, pp. 217-223.
Virmani, V, (2004), “Estimating the output gap for the Indian economy: Comparing
results from unobserved components models and the Hodrick-Prescott
filter”, Working Paper No 2004-04-02, Indian Institute of Management,
Ahmedabad.
20
Walsh, J.P. (2011),. "Reconsidering the Role of Food Prices in Inflation," Working
Paper No. 11/71, International Monetary Fund, Washington, April.
Woodford, M., (2003), Interest and prices: Foundations of a theory of monetary
policy. Princeton University Press.
21
Appendix 1
Table 1. Estimated Open-Economy Phillips Curve for Period 2 (Q2 2004 – Q4
2009)
OLS IV Reg 2SLS
Time period 2 Q2 2004 –Q4 2009
0.8 0.9 0.8 0.8
(0.1) *** (0.4)*** (0.1)*** (0.1)***
0.7 0.9 1.5 2.0
(0.4) ** (0.09) *** (0.7)** (1.0)**
-1.7 -2.2 -2.6 -3.4
(0.6)*** (0.6) *** (0.95)*** (1.3)***
-0.08 -0.05
(0.04) ** (0.03)*
-0.04 -0.05
(0.02) *** (0.02)***
C 1.4 1.1 1.5 1.4
(0.6)*** (0.59)** (0.6)*** (0.7)***
Number of Observations 23 23 23 23
Adj. R-squared 0.92 0.92 0.81 0.92
Durbin-Watson Statistic 2.3 2.6 2.0 2.3
Note:
1) Standard errors are given in parentheses.
2) *, **, *** indicate significance at the 10 %, 5 %, and 1% level, respectively.
3) We use one quarter lag of inflation ( ) as a proxy for expected inflation.
1t
)ˆˆ( YYn
tt
sdumYYn
tt*)ˆˆ(
1t
22
Table 2. Data Definitions and Sources
Variables
Name
Definition Data Source
Year-over-year percentage
change in the composite
consumer price index
(CCPI).
To construct the CCPI, we have followed
the methods used by Singh, B.K. and
Joseph, M. (2009). Data on consumer
price indices of the four income groups
(agricultural labourers, rural labourers,
industrial workers and urban non-
manual employees) has been taken from
the Online Database on the Indian
Economy, Reserve Bank of India.
Deviation of actual output
( ) from potential output
( ), derived using the
Kalman filter method.
Authors’ own calculations using
quarterly real GDP data from the Online
Database on the Indian Economy,
Reserve Bank of India.
Supply shock dummy,
which takes the value 1 for
the periods of supply
shocks; otherwise, it is
equal to 0.
Authors’ own calculations, crosschecked
with monthly publications of the RBI
Bulletin (1997 - 2009),
Year-over-year percentage
change in the trade-
weighted real effective
exchange rate, based on
the CPI
Karan Singh, B. and Mathew, Joseph,
Why Are Trade Deficits Worsening in the
Case of India? (Unpublished)
Year-over-year percentage
change in the trade-
weighted real effective
exchange rate, based on
the WPI
Online Database on the Indian Economy,
Reserve Bank of India.
t
YYn
ttˆˆ
Y tˆ
Yn
tˆ
tsdum
Working Paper No. 236
INDIAN COUNCIL FOR RESEARCH ON INTERNATIONAL ECONOMIC RELATIONS
LATEST ICRIER’S WORKING PAPERS
NO.
TITLE
AUTHOR YEAR
258
TOWARD A FRAMEWORK FOR
IMPLEMENTATION OF CLIMATE
CHANGE TREATY THROUGH SELF-
ENFORCING MECHANISMS
MEETA KESWANI
MEHRA
SAPTARSHI MUKHERJEE
MONICA DUTTA
JANUARY
2012
257
INDIA’S EXPERIENCE IN
NAVIGATING THE TRILEMMA: DO
CAPITAL CONTROLS HELP?
R. KOHLI JUNE 2011
256
MONETARY POLICY AND CREDIT
DEMAND IN INDIA AND SOME EMES
B L PANDIT
PANKAJ VASHISHT
MAY 2011
255
ENHANCING INTRA-SAARC TRADE:
PRUNING INDIA’S SENSITIVE LIST
UNDER SAFTA
NISHA TANEJA
SAON RAY
NEETIKA KAUSHAL
DEVJIT ROY
CHOWDHURY
APRIL 2011
254
FINANCIAL TRANSACTIONS TAXES
PARTHASARATHI
SHOME
APRIL 2011
253
INDIAN ECONOMY: SELECTED
METHODOLOGICAL ADVANCES
MATHEW JOSEPH
KARAN SINGH
RANJAN KUMAR DASH
JYOTIRMOY
BHATTACHARYA
RITIKA TEWARI
FEBRUARY
2011
252
HEALTHCARE DELIVERY AND
STAKEHOLDER’S SATISFACTION
UNDER SOCIAL HEALTH
INSURANCE SCHEMES IN INDIA: AN
EVALUATION OF CENTRAL
GOVERNMENT HEALTH SCHEME
(CGHS) AND EX-SERVICEMEN
CONTRIBUTORY HEALTH SCHEME
(ECHS)
SUKUMAR VELLAKKAL
SHIKHA JUYAL
ALI MEHDI
DECEMBER
2010
251
ADDRESSING NEW SERVICE
SECTORS IN WTO/FTAS: EXPRESS
DELIVERY AND INDIA
ARPITA MUKHERJEE
PARTHAPRATIM PAL
RAMNEET GOSWAMI
AUGUST
2010
250
SPORTS RETAILING IN INDIA:
OPPORTUNITIES, CONSTRAINTS
AND WAY FORWARD
ARPITA MUKHERJEE
RAMNEET GOSWAMI
TANU M GOYAL
DIVYA SATIJA
JUNE 2010
249
THE SERVICE SECTOR AS INDIA’S
ROAD TO ECONOMIC GROWTH?
BARRY EICHENGREEN
POONAM GUPTA
APRIL 2010
1
Working Paper No. 236
INDIAN COUNCIL FOR RESEARCH ON INTERNATIONAL ECONOMIC RELATIONS
About ICRIER
Established in August 1981, ICRIER is an autonomous, policy-oriented,
not-for-profit economic policy think tank. ICRIER's main focus is to
enhance the knowledge content of policy making by undertaking
analytical research that is targeted at improving India's interface with the
global economy. ICRIER's office is located in the prime institutional
complex of India Habitat Centre, New Delhi.
ICRIER’s Board of Governors comprising leading policy makers,
academicians, and eminent representatives from financial and corporate
sectors is presently chaired by Dr. Isher Ahluwalia. ICRIER’s team is led
by Dr. Parthasarathi Shome, Director and Chief Executive.
ICRIER conducts thematic research in the following seven thrust areas:
Macro-economic Management in an Open Economy
Trade, Openness, Restructuring and Competitiveness
Financial Sector Liberalisation and Regulation
WTO-related Issues
Regional Economic Co-operation with Focus on South Asia
Strategic Aspects of India's International Economic Relations
Environment and Climate Change
To effectively disseminate research findings, ICRIER organises
workshops, seminars and conferences to bring together policy makers,
academicians, industry representatives and media persons to create a
more informed understanding on issues of major policy interest. ICRIER
invites distinguished scholars and policy makers from around the world to
deliver public lectures on economic themes of interest to contemporary
India.
Recommended